Addition of unactivated thiols to epoxides and oxetanes catalyzed by a rhenium-oxo complex

Article abstract of DOI:10.1016/j.catcom.2014.01.005

A method for synthesizing β- and γ-hydroxy thioethers via addition of unactivated thiols to epoxides and oxetanes has been developed. This reaction is proposed to proceed through an unconventional mode of activation of the heterocycles. The transformation is catalyzed by ReO₂I(PPh ₃)₂ affording β- and γ-hydroxy thioethers in moderate to excellent yield with excellent regioselectivity.

Full text of DOI:10.1016/j.catcom.2014.01.005

Catalysis Communications 47 (2014) 67–70
Contents lists available at ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Short Communication
Addition of unactivated thiols to epoxides and oxetanes catalyzed by a
rhenium-oxo complex
⁎
Alexandra D. Badiceanu, Alyson E. Garst, Meredith E. Trubitt, Kristine A. Nolin
Department of Chemistry, University of Richmond, Richmond, VA 23173, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A method for synthesizing β- and γ-hydroxy thioethers via addition of unactivated thiols to epoxides and
oxetanes has been developed. This reaction is proposed to proceed through an unconventional mode of activa-
tion of the heterocycles. The transformation is catalyzed by ReO2I(PPh3)2 affording β- and γ-hydroxy thioethers
in moderate to excellent yield with excellent regioselectivity.
Received 16 December 2013
Accepted 6 January 2014
Available online 14 January 2014
© 2014 Elsevier B.V. All rights reserved.
Keywords:
Rhenium catalysis
Epoxides
Oxetanes
Regioselective
Thiols
Regio-, stereo-, and chemoselective bond forming reactions are the
key to efficient syntheses of organic compounds and materials. A fre-
quently employed strategy for achieving selective reactivity is the use
of transition metal catalysis. Often, transition metals can provide access
to reaction pathways that may not be available to main group elements.
One privileged class of transition metal catalysts is the metal-oxo com-
plexes, which contain metal–oxygen multiple bonds [1]. The use of
metal-oxo complexes as catalysts in the oxidization of unsaturated hy-
drocarbons to yield epoxides, diols, and carbonyl compounds is well
documented [2]. Espenson and Gable have elegantly shown that, in
the case of epoxidations catalyzed by high oxidation state rhenium-
oxo complexes, the oxygen atom transfer is reversible [3]. They discov-
ered that equilibrium is established between the epoxide and the corre-
sponding olefin and proposed that the interconversion of these two
compounds proceeded via the formation of a rhenium diolate I
(Eq. (1)). The formation of I from the epoxide implies a variation from
the traditional Lewis acidic activation of epoxides.
The deoxygenations of epoxides using rhenium-oxo complexes
were rendered catalytic with the addition of triphenylphosphine as an
oxygen acceptor [4]. We postulated that this non-traditional mode of
activation could be incorporated into catalytic nucleophilic ring opening
reactions by intercepting diolate I with a nucleophile before deoxygen-
ation. These reactions would be complementary to existing Lewis acid
reactions [5] and allow for controlled additions to strained heterocycles.
Based on the results reported by Gable and coworkers in the deoxygen-
ation reaction described above, we began examining rhenium(V)-oxo
complexes in the catalytic activation of epoxides toward the addition
of thiols [6]. It was envisioned that, after activation of the epoxide,
diolate intermediate I could be intercepted with the addition of a thiol
(Scheme 1). Nucleophilic substitution on the diolate and proton transfer
would result in the formation of alkoxy rhenium hydroxide complex II.
Tautomerization regenerates the metal-oxo ligand [7] and the cycle is
closed after liberation of the functionalized product [8].
A number of rhenium(V)-oxo complexes were examined as poten-
tial catalysts in the reaction of 4-methylbenzenethiol with styrene
oxide, 8 (Eq. (2)). A representative sampling, 1–7, is shown in Fig. 1.
Measureable amounts of the desired β-hydroxy thioether 9 were ob-
tained in the presence of catalytic quantities of 1–7; however, these re-
actions also showed significant amounts of styrene, 12, via the
competitive deoxygenation pathway. Also present in the reaction mix-
tures was the hydration product 1-phenylethanediol, 10. The reactions
promoted by complexes 1–5 also lead to the formation of 2-chloro-2-
phenylethanol, 11, presumably through transfer of the chloride ligand
from the metal. A similar iodinated product was not observed for reac-
tions run in the presence of 6 or 7 [9]. Significant degradation of the
starting material to unidentified, possibly polymeric byproducts was
observed in the presence of 6. Reactions catalyzed by 7 produced the
ð1Þ
⁎
Corresponding author at: 28 Westhampton Way, Department of Chemistry,
University of Richmond, Richmond VA 23173, USA. Fax: +1 804 287 1897.
E-mail address: knolin@richmond.edu (K.A. Nolin).
1566-7367/$ – see front matter © 2014 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.catcom.2014.01.005

68
A.D. Badiceanu et al. / Catalysis Communications 47 (2014) 67–70
Table 1
Optimization studies.
Scheme 1. Proposed catalytic cycle for addition of thiols to epoxides.
highest conversion of starting material to 9 with 10 and 12 as the most
notable byproducts. Therefore, reaction optimization was continued,
employing 7 as the catalyst.
Increasing the concentration allowed for a further increase in the yield
of 9 relative to diol 10 and decrease in the reaction times (entries
7–8). In addition, the catalyst loading could be decreased to 1.0% with
negligible impact on the yield; however, extended reaction times
were required (entry 9).
The reactivity of styrene oxide toward the addition of substituted
thiophenols derivatives was probed for generality. The reactions were
performed under the optimized conditions: 2.0 M in dry ethyl acetate
at room temperature in the presence of 1 mol% of 7. The reactions
were quenched upon completion as determined by TLC (approximately
4–8 h). Electron-rich and electron deficient thiophenol derivatives were
added to styrene oxide to produce the corresponding β-hydroxy
thioethers in good yield (62–77%, Table 2, entries 1–4). For these
reactions, only one regioisomer was observed. The observed regioselec-
tivity, addition of the thiols to the benzylic position of styrene oxide, is
attributed to the stabilizing ability of the aromatic ring toward charges
developing in the transition state leading to the thiol addition. The re-
gioselectivity was preserved with the reaction of styrene oxide deriva-
tives 12 and 13 with 4-methylbenzenethiol (entries 5 and 6).
The reactivity of alkyl and benzyl substituted epoxides was exam-
ined for reactivity under the above described reaction conditions with
a reaction time of 4–24 h. 1,2-Hexene oxide, 17, was highly reactive to-
ward the addition of electron-rich and electron-deficient thiophenol de-
rivatives (49–95%, Table 3, entries 1–4). The reaction proceeded with
excellent regioselectivity leading to the formation of the secondary alco-
hol. We postulate that regioselectivity is dictated by minimization of
steric repulsion of the incoming thiol leading to opening of alkyl-
ð2Þ
For the optimization studies, the addition of 1 equivalent of 4-
methylbenezenethiol to styrene oxide in the presence of 7 was exam-
ined. The percent conversion and product distribution [10] were deter-
mined by 1H NMR versus an internal standard. In the presence of 5 mol%
of 1 at room temperature in CH2Cl2 (0.2 M), styrene oxide was trans-
formed to a mixture of 9, 10, and 12 with a total conversion of 70%
(Table 1, entry 1). Highly coordinating acetonitrile permitted very little
reactivity with only 8% conversion of styrene oxide to 9 and 10 with no
deoxygenation observed (entry 2). In hexanes, a decreased conversion,
relative to CH2Cl2, was obtained; however, there was a notable decrease
in hydration and deoxygenation products (entry 3). Gratifyingly, ethyl
acetate showed complete conversion of the starting material to 9, the
major product; however, the reaction was prone to hydrolysis (entry
4). By using ethyl acetate that had been dried over activated molecular
sieves, the relative amount of thioether 9 formed was markedly in-
creased. Comparable yields were observed when the reaction was con-
ducted at lower temperatures with extended reaction times (entry 6).
Fig. 1. Sampling of re-oxo complexes screened for catalytic activity.

A.D. Badiceanu et al. / Catalysis Communications 47 (2014) 67–70
69
Table 2
have shown that the opening of cyclic epoxides will proceed through
the axial attack of the nucleophile [11]. Based on this precedence,
regioisomer 31a would be formed from the opening of cis-(+)-
limonene oxide (cis-30) and 31b from trans-(+)-limonene oxide
(trans-30). The initial cis:trans ratio of (+)-limonene oxide was
1.0:1.3. The enriched regioselectivity of the addition products, 1.0:1.9,
relative to the starting ratio of the cis:trans isomers, 1.0:1.3, suggests
that an isomerization of the epoxide was taking place. To test this, a
sample of (+)-limonene oxide was reacted with 7 in the absence of
thiol. After 24 h, the reaction was quenched and the ratio of the diaste-
reomers was determined by 1H NMR and found to be a 1.0:3.1 mixture
of cis:trans isomers (Eq. (4)). An isomerization to trans-30 proceeding
concurrently with the addition reactions would enable for an enrich-
ment of 31b in the product distribution. It is envisioned that such an
isomerization could be accomplished through a deoxygenation/epoxi-
dation sequence. Such a process would mirror the observations of
Gable and coworkers.
Reaction of styrene oxides with thiophenol derivatives.
substituted epoxides at the less substituted position. Under standard
conditions, the opening of benzylic epoxide 18 also proceeded in good
to excellent yield, 56–93%, with excellent regioselectivity (Table 3 en-
tries 5–8). With both epoxides, measurable amounts of the minor
regioisomer were formed from the addition of the more electron rich
thiophenol derivatives (entries 1, 5, and 6).
The scope of the reaction was further probed for the tolerance of var-
ious functional groups. Moderate to good yields were obtained from the
reaction of epoxides bearing α-alkoxy substitution (56–81%, Table 4,
entries 1 and 2). The presence of the phenoxy functionality did not in-
fluence the regioselectivity of the reaction; the only regioisomer obtain-
ed resulted from the addition to the less substituted position of the
epoxide. However, glycidyl butyrate, 22, was reacted to form a
15.8:1.0 inseparable mixture of the regioisomers. This addition reaction
is not limited to mono-substituted epoxides. Tetra substituted epoxide
23 provided tertiary thioether 27 in moderate yield (42%, entry 3).
The reaction of 1,2-epoxy-9-decene, 24, with 4-chlorothiophenol pro-
vided an 85% yield of an 86:14 mixture of regioisomers with β-hydroxy
thioether 28 as the major product (entry 4). Higher yields, 91%, were
obtained when 24 was reacted with 4-methoxythiophenol. Thioether
29 was obtained as a 7.1:1.0 mixture of regioisomers (entry 5). Gratify-
ingly, no oxidation of the terminal alkene was observed in either
reaction.
ð3Þ
ð4Þ
Having successfully activated epoxides, we looked to extend this
methodology to the addition of thiols to oxetanes. It was expected
that oxetanes could also be activated in a manner analogous to the pos-
tulated mechanism (Scheme 1) to form a six-membered diolate inter-
mediate. Gratifyingly, 2-phenyloxetane underwent addition of 4-
clorothiophenol in the presence of 1 mol% of 7 under the conditions
used for the epoxide reactions. A brief optimization study identified
that a 0.5 M reaction solution in dry CH2Cl2 produced the highest con-
version of 32 to the thioether 33. One regioisomer of thioether 33 was
isolated in a 73% yield (Eq. (5)). It is of note that no polymerization of
the oxetane was observed [12].
The addition of 4-chlorothiophenol to (+)-limonene oxide, 30, pro-
vided our first insight into the mechanism of the reaction. The reaction
produced a 1.0:1.9 mixture of 31a and 31b as single diastereomers of
each regioisomer with an overall yield of 77% (Eq. (3)). Past studies
Table 3
Addition of thiols to alkyl and benzyl epoxides.
ð5Þ
Rhenium(V)-oxo complex 7 was shown to be an effective catalyst
for the addition of unactivated thiols to differentially substituted epox-
ides and 2-phenyloxetane. The reaction proceeded in moderate to ex-
cellent yield with excellent regioselectivity. Elaboration of this
reaction and elucidation of the mechanism are underway in our labora-
tory and will be reported in due time.

70
A.D. Badiceanu et al. / Catalysis Communications 47 (2014) 67–70
Table 4
Extension of substrate scope.
(b) M. Wu, E.J. Jacobsen, Org. Chem. 63 (1998) 5252–5254;
Acknowledgements
(c) (1)W. Su, J. Chen, H. Wu, C. Jin, J. Org. Chem. 72 (2007) 4524–4527.
[6] (a) Some recently reported methods for adding thiols to epoxides. B. Das, P.
Balasubramanyam, M. Krishnaiah, B. Veeranjaneyulu, D. Sudhakar, Synth.
Commun. 40 (2010) 2113–2121;
We gratefully acknowledge the University of Richmond, the Thomas
F. and Kate Miller Jeffress Memorial Trust, and the Research Corporation
for Science Advancement for their financial support including student
summer research fellowships.
(b) R. Rani, S. Pattanayak, J. Agarwal, R.K. Peddinti, Synth. Commun. 40 (2010)
2658–2666;
(c) M. Shailaja, A. Manjula, B. Vittal Rao, Synth. Commun. 40 (2010) 3629–3639;
(d) S. Eagon, N. Ball-Jones, D. Haddenham, J. Saavedra, C. DeLieto, M. Buckman, B.
Singaram, Tetrahedron Lett. 51 (2010) 6418–6421;
(e) D. Lanari, R. Ballini, S. Bonollo, A. Palmieri, F. Pizzo, L. Vaccaro, Green Chem. 13
(2011) 3181;
References
(f) M.M. Mojtahedi, M.S. Abaee, A. Rajabi, P. Mahmoodi, S. Bagherpoor, J. Mol. Catal.
A: Chem. 361–362 (2012) 68–71;
[1] W.A. Wugent, Nugent, J.M. Mayer, Metal–Ligand Multiple Bonds, Wiley, New York,
1988.
[2] In: B.M. Trost, I. Fleming (Eds.), Comprehensive organic synthesis, vol. 7, Pergamon
Press, New York, 1991, (and reference therein).
[3] K. Gable, J. Juliette, J. Am. Chem. Soc. 118 (1996) 2625–2633;
(b) K. Gable, F. Zhuravlev, J. Am. Chem. Soc. 124 (2002) 3970–3979;
(c) K. Gable, E. Brown, J. Am. Chem. Soc. 125 (2003) 11018–11026;
(d) K.P. Gable, K. Khownium, P. Chuawong, Organometallics 23 (2004) 5268–5274;
(e) Z. Zhu, A.M. Al-Ajlouni, J.H. Espenson, Inorg. Chem. 35 (1996) 1408–1409.
[4] K. Gable, F. Zhuravlev, A. Yokochi, Chem. Commun. (1998) 799–800;
(b) K.P. Gable, E.C. Brown, Synlett (2003) 2243–2245.
(g) S.R. Lanke, B.M. Bhanage, Catal. Commun. 41 (2013) 1–5.
[7] Such a mechanism may be operational in the reported methyltrioxorhenium (MTO)
catalyzed one-pot epoxidation cyclication of 5-hydroxyalkenes to yield tetrahydro-
furans. H. Tan, J.H. Espenson, J. Mol. Catal. A Chem. 152 (2000) 83–89.
[8] Similar mechanism proposed for the hydrolysis of epoxides using MoO4 salts. J.R.
Briggs, A.M. Harrison, J.H. Robson, Polyhedron 5 (1986) 281–287.
[9] Studies are ongoing to determine the possible presence of an iodinated intermediate.
[10] Distribution percentages that do not add up 100% indicate the presence of some ad-
ditional, unidentified byproduct(s).
[11] C.M. Byrne, S.D. Allen, E.B. Lobkovsky, G.W.J. Coates, Am. Chem. Soc. 126 (2004)
11404–11405.
[12] H. Kudo, T.J. Nishikubo, Polym. Sci. Polym. Chem. 45 (2007) 709–726.
[5] (a) Examples of Lewis acid catalyzed additions of thiols to epoxides. B.C. Ranu, T.
Can Mandal, J. Chem. 84 (2006) 762–770;